Alaoui/Phillips - UC San Franscico

Principal Investigators: Hassan Alaoui, Ph.D. & Joanna Phillips, M.D., Ph.D.
Institution: University of California, San Franscico, CA

Our project, Extracellular Sulfatases As Novel Biomarkers For Early Detection of Cancer, is a collaborative project involving the laboratories of Hassan Alaoui (PI), Joanna Phillips (PI), Steven Rosen (Co-PI), and Annette Molinaro (Co-PI). Our goal is to determine the clinical utility of plasma levels of extracellular heparan sulfate endosulfatases (SULFs) or their substrates, as robust biomarkers for the early detection of malignant lung and brain cancer.

In the Alaoui lab, our work is focused on discovering novel targeted therapies and diagnostic biomarkers with clinical utility for the treatment of lung cancer and other thoracic malignancies, based upon a better understanding of the molecular mechanisms pertinent to these illnesses. Lung cancer is the most frequently diagnosed cancer and the leading cause of cancer death worldwide. There is great interest in the mechanisms that lead to the uncontrolled growth of lung cells and the formation of tumors. Scientists are therefore interested in the chemical signals that cause cancer cells to grow. We are studying two proteins, called SULF1 and SULF2, which are responsible for controlling the quantity of these signals available on the outside of cells and thus exert control over cell growth. Our recent study provides the first evidence that SULF2 promotes human lung carcinogenesis by regulation of Wnt signaling and the kinase activities of three critical receptors (i.e., EGFR, IGF-1R and cMet). Dysregulation of each of these three receptors has been causally linked to lung cancer development, increased resistance to chemotherapies, and lung cancer progression. SULF1 and SULF2 are enzymes that are secreted and act on heparan sulfate proteoglycans (HSPGs) in the extracellular matrix. Our lab is currently investigating these enzymes as new therapeutic targets and potential biomarkers for early detection of cancer. Lung cancer is one of the most deadly cancers, and, in those with the most advanced disease, median survival from diagnosis is less than one year. Earlier diagnosis of disease using a robust biomarker would improve survival. Perhaps this is most evident in non-small cell lung cancer where early stage detection of lung cancer can lead to early interventions thus enhancing probability for survival. We are testing the clinical utility of the SULFs as biomarkers of lung cancer. As secreted enzymes, the SULFs are present in the extracellular environment and have great potential as novel biomarkers for early detection of cancer in bodily fluids. We have found that the SULFs genes are significantly upregulated in lung tumors from human patients. In addition, we have shown that either one or both SULFs can be detected in the plasma of a subset of newly diagnosed, early stage NSCLC patients before surgical resection.

In the Phillips lab we are interested in understanding how invading brain tumor cells interact with components of the tumor microenvironment and how these key interactions can be targeted for therapy and used as potential biomarkers of disease. In our studies, we have identified SULF2 production as a novel mechanism by which glioblastoma (GBM), the most common and most malignant type of primary brain cancer, alters HSPGs to promote receptor tyrosine kinase signaling and drive tumor growth.We have demonstrated that primary human GBM express SULF2, and by decreasing SULF2 levels in tumors we have decreased tumor growth, prolonged host survival, and demonstrated a decrease in activation of PDGFRalpha and downstream signaling pathways. In addition, we have shown that SULF2 can be detected in the blood of mice harboring human tumor xenografts. As our data suggest that SULF2 and alterations in HSPGs play a functional role in brain cancer, we are excited to develop assays that will allow us to detect these alterations in blood as a biomarker for the early detection of disease.

In our Alliance project, we have assembled an exceptional team of scientists and clinicians to assess the use of the SULFs and their activities as biomarkers for malignant lung and brain cancer. Our team includes scientists with expertise in SULFs studies, glycobiology, biostatistics, and physicians who treat patients with lung and brain cancer and have access to critical clinical specimens.

Synopsis of Research and Network Resources

Introduction: A Brief Summary of the State of the Science and Research Needs

Lung cancer and brain cancer are two of the most deadly cancers, and, in those with the most advanced disease, median overall survival from diagnosis is less than one year. It is widely accepted that earlier detection and better patient stratification would improve survival. Perhaps this is most evident in non-small cell lung cancer (NSCLC) where early stage detection of lung cancer can increase the 5-year overall survival from 15% for all stages to 80% for stage IA [1]. However, currently there are no clinical biomarkers for the early detection of lung or brain cancer. Hence, novel lung and brain cancer biomarkers are urgently needed. Ideally, identification of these biomarkers would be based upon a mechanistic rationale and functional importance in tumorigenesis [2]. The extracellular endosulfatases (SULF1 and SULF2) are overexpressed in a wide assortment of human cancers [3-6] and SULF2, in particular, has been implicated, by our group and others, as a driver of carcinogenesis in NSCLC [7], pancreatic carcinoma [8], malignant astrocytoma [4,9,10], and hepatocellular carcinoma [11]. As these extracellular enzymes that are both tethered to the cell membrane and secreted, the SULFs and their heparan sulfate proteoglycan (HSPG) substrates are present in the extracellular environment. HSPGs carry out enumerable signaling functions by using their sulfated chains to bind diverse protein ligands, such as growth factors, morphogens, and cytokines. These interactions depend on the pattern of the sulfation [5,12,13]. The 6-O-sulfation (6OS) of glucosamine is known to be key for many ligand interactions and may be altered in cancer [14]. SULFs selectively remove the critical 6OS modifications?from HSPGs, and in so doing, prevent ligand sequestration. We hypothesize that the blood levels of SULF1 and SULF2 or their HSPG substrates can serve as biomarkers for the early detection of NSCLC and malignant astrocytoma.

Laboratory-specific studies to meet the research needs

The overall goal of our proposal is to determine whether blood levels of SULF1 and SULF2 or their HSPG substrates can serve as biomarkers for the early detection of NSCLC and malignant astrocytoma. To achieve this objective, we are pursuing the following strategies:

A. Using human biospecimens from lung and brain tumor patients, including tumor tissue and blood, we will define the patterns and extent of SULF expression as a function of tumor stage for NSCLC and grade for astrocytoma. We will then determine SULF levels in blood (plasma or serum) of cancer patients and age- and gender-matched healthy individuals, and correlate these levels with tumor presence and progression. To assess the blood levels of SULFs (SULF1 and SULF2) we have devised sandwich ELISAs for the detection of either SULF1 or SULF2 in human blood, using our in-house mAbs specific for SULF1 or SULF2 (with no cross-reactivity for SULF2 or SULF1, respectively). We have shown that both of these assays are applicable to determine SULF levels in serum and plasma.

B. Assays for SULF bio-activity would improve specificity of the SULFs biomarker approach; We hypothesize that increased SULF activity in tumor tissue will result in decreased levels of 6OS on HSPGs shed into blood and increased blood levels of heparin-binding factors mobilized by the SULFs. These could potentially be used as functional blood biomarkers for NSCLC and malignant astrocytoma. Thus, we will also determine the patterns and extent of 6OS modification on HSPGs and the levels of heparin-binding factors in the blood of NSCLC and astrocytoma patients in age and gender matched healthy individuals, and correlate with tumor presence and progression.

1. To determine the 6OS status on select HSPGs from blood, we will develop a capture ELISAs to quantify the 6OS status of select HSPGs that are known to be elevated in the blood of NSCLC and malignant astrocytoma patients, including Syndecan 1 (SDC1). SDC1 is a prominent HSPG, which is proteolytically shed from the plasma membrane. SDC1 levels are elevated in malignant astrocytoma [15] and in NSCLC [16]. Furthermore, SDC1 is present in normal blood and markedly increases in the blood of NSCLC patients. Additionally, high blood levels of SDC1 correlated with the presence of large tumor masses in NSCLC patients [16], suggesting that the origin of the elevated SDC1 in the blood may be due to shedding of the protein from the tumor mass. Importantly, SDC1 has been demonstrated to be a HSPG substrate for the SULFs [17,18]. Therefore, it is plausible that the sulfation status of SDC1 in patient blood samples may reflect the action of overexpressed SULFs in the tumor and thus would show an altered sulfation pattern, i.e. reduced levels of 6OS. To determine the 6OS sulfation status of SDC1 in blood we have recently developed a capture ELISA binding the SDC1 from blood via a coated SDC1 antibody on plastic wells and detection of sulfation status of the bound SDC1 with a biotinylated growth factor (such as VEGF, FGF-1) whose binding to heparan sulfate requires 6OS [19]. We are also using a second ELISA to determine the overall level of SDC1 protein in the blood, so that the level of 6OS could be normalized to the level of the protein. A similar strategy could be used to evaluate the sulfation status of a range of other HSPGs in the plasma of cancer patients vs. age- and gender-matched healthy controls.
    
2. To characterize the blood levels of heparin-binding factors (growth factors, morphogens, cytokines, chemokines) potentially mobilized by SULF action in the tumor, we will develop new ELISAs to measure the levels of SULF mobilized ligands in blood. SULFs have a proven ability to mobilize a number of protein ligands (including VEGF, FGF-1, CXCL12 (SDF-1), CCL21, IL-8, and IP-10) from their pre-association with heparin/HSPGs [19]. Therefore, a consequence of SULF overexpression in a tumor could be the release of HSPG-sequestered factors and increased levels in the blood. In this regard, it is notable that a number of heparin-binding factors (VEGF, IL-8, IL-6, BMP-2, amphiregulin, HGF) are either elevated in the blood of NSCLC patients vs. controls, increased in blood levels with disease progression, or show positive correlations between blood levels and disease prognosis [20,21]. Moreover, dozens of other known heparin/HSPG-binding growth factors, chemokines and cytokines are detectable in normal blood by ELISA [22,23] and are yet to be investigated in NSCLC and astrocytoma patient samples. The central question is whether the blood levels of these factors could be used alone or in combination with SULF/HSPG 6OS levels as signatures for early detection of cancer. To measure the blood levels of HS-binding factors including those known to be elevated in NSCLC (VEGF, IL-8, IL-6, BMP-2, amphiregulin, HGF) [20,21] and malignant astrocytoma (IGFBP3, MMP9, VEGF, YKL40, and angiopoietins-1 and -2 [24-26], we have recently developed quantitative ELISAs based on heparin-BSA capture of these ligands from blood followed by detection with biotinylated monoclonal antibodies specific for each individual ligand. The absolute blood levels of these heparin sulfate binding ligands are determined using a standard recombinant protein for each factor. We will use these ELISAs to evaluate the blood levels of these ligands in NSCLC and astrocytoma patients as a function of disease stage and grade. Importantly, these will be compared to age- and gender-matched healthy controls and control patients with non-neoplastic or benign pulmonary or neurologic disease. To determine whether the plasma levels of these factors correlate with tumor cell expression of SULF, we will compare their plasma levels in a cohort of patients with the SULFs levels assessed in paired tumor samples taken from the same patients. This data will be used to determine sensitivity and specificity of blood factors to predict SULF-positive disease. Collectively, our proposed research will test the hypothesis that the extracellular sulfatases, in addition to being drivers of tumorigenesis, can serve as biomarkers for the early detection of malignant lung and brain cancer. These investigations build upon the mechanistic role of the SULFs in tumorigenesis, and address the currently unmet need for biomarkers for the early detection of lung and brain cancer. Biomarker assays based on the SULFs and SULF bioactivity could have a profound impact on clinical care and prognosis of patients with these deadly cancers.

Resources and Reagents for Sharing

1. DNA: cDNAs for human and murine SULFs (HSulf-1, HSulf-2, MSulf-1, MSulf-2) in expression vector pcDNA3.1/Myc-His. C87A and C88A inactive mutants of each. pIZ lentiviral plasmids encoding HSulf-1 and HSulf-2 both active forms and inactive mutants. Lentiviral knockdown plasmids (pLVTHM vector) with SULF knockdown shRNAs and control shRNA.
    
2. Antibodies: 18 SULF2-specific monoclonal antibodies (mAbs): 17 SULF1-specific mAbs. Cell lines: Human SULF1- or SULF2-positive NSCLC and astrocytoma cell lines. Hybridomas for the SULF1 and SULF2 mAbs mentioned above. HEK-293 cells with stable-overexpression of either HSULF1 or HSULF2.
    
3. Other reagents: heparin-BSA.
    
4. Screening assays: Sandwich ELISAs for detection of SULFs in the blood. Sandwich ELISAs for heparin bound-ligands detection in which ligands are captured with heparin-BSA and detected by antibodies specific to each individual ligand. (1) Endosulfatase assay in which the substrate, heparin-BSA, is coated onto plastic ELISA wells. SULF is incubated in the wells and the altered sulfation pattern on the heparin is detected with a biotinylated growth factor (such as VEGF, FGF-1) whose binding to heparan sulfate requires 6OS. (2) Sandwich ELISAs for SDC1 detection in the blood. (3) Sandwich ELISAs to evaluate the 6OS sulfation status of the shed SDC1 in blood where SDC1 is captured from blood via a coated SDC1 antibody on plastic wells and sulfation status of the bound SDC1 is detected with a biotinylated heparin binding growth factor, as described above.
    
5. Clinical assets: Access to critical clinical specimens (archived and prospective) to include flash-frozen and formalin-fixed human tumor biospecimens, as well as corresponding blood samples from patients with lung or brain cancer before surgery; Blood samples from age- and gender- matched healthy individuals.

Public Health Implications and Advancing the Field of Glycobiology

Lung and brain cancer are significant public health problems in the U.S. and worldwide. Most patients with these cancers are diagnosed at latestages of the disease and the prognosis is particularly dismal, with mean survival rates of less than two years from diagnosis. Biomarkers for earlier diagnosis of these cancers are urgently needed as these would improve survival and reduce morbidity. Our primary focus is the development of glycomics-based diagnostic tests for the early detection of these two lethal cancers, and thus have the potential to have major public health implications. Once developed, tumor-specific blood biomarkers would also be useful for determining response to therapy and for the detection of recurrent disease. Specifically, a blood biomarker test based on the SULFs, with a demonstrated mechanistic rationale, could greatly improve disease diagnosis and management, including: earlier diagnosis of cancer, a decrease in false-positive and false-negative test results, reduction in unnecessary diagnostic procedures, more rapid feedback on response to therapy, and overall provide more cost-effective management of these deadly illnesses. In addition, as our studies are mechanistically based, they have the potential to help unravel how glycosaminoglycan alterations may drive disease onset and progression.

Cancer-Specific Relevance: detection, prevention and treatment

Our studies focus on a target that has been implicated as a driver of tumorigenesis in many cancers including lung, pancreas, and brain. There is a great need for new diagnostic tests in lung and brain cancers, because prognosis for patients with early stage, localized cancer is much better than for those with established and disseminated disease. Development of robust biomarker assays based on the SULFs and SULF bioactivity, either as stand-alone tests or in combination with assays that measure the levels of other circulating biomarkers such as CYFRA could have a profound impact on clinical care and prognosis of patients with these cancers. These blood tests may provide sufficient specificity to complement the imaging modalities routinely used in population screening for cancer. Particularly, these tests could be used widely in high-risk populations, including those with a family history of cancer, patients with cancer predisposition syndromes, smokers and former smokers, veterans whose smoking history and exposure to other lung carcinogens during active duty service has increased their risk for cancer, and patients in remission from cancer.

Moreover, we and others have found that higher expression of SULF2 protein in esophageal tumors and hepatocellular carcinoma are associated with worse prognosis. Additionally, SULF2 methylation has been shown to have a prognostic value for survival and increased sensitivity to topoisomerase-I inhibitors in both lung and gastric cancers [27,28]. Accordingly, a screening ELISA for SULF2 levels may help with patients’ stratification and create a molecular-based test to guide treatment decisions for these cancer subtypes.

Opportunities for Collaboration

As detailed above, we have developed a set of ELISAs for the detection of blood biomarkers and we have generated novel monoclonal antibodies against enzymes that modify glycoproteins. Additionally, our team of investigators and collaborators have extensive expertise in pathology, testing and validation of assays for clinical use, and the treatment of lung and brain cancer patients. Through the UCSF Thoracic Oncology Program and UCSF Brain Tumor Research Center Biorepository, we have access to a large number of archived and prospective biospecimens (described above) from patients with lung or brain cancer and controls. These precious human biospecimens would be available for future trans-Alliance Collaborations to investigate and screen lung and brain cancer samples and blood from these patients for potential novel biomarkers. In addition, the SULFs are also overexpressed in diverse cancers including hepatocellular, breast, head and neck, gastric, and pancreatic carcinomas and multiple myeloma [3,5]. Thus, our studies could have implications for many cancers, and we look forward to sharing our expertise and resources for joint trans-Alliance activities. These studies could further elucidate pathways critical for tumor progression and potentially identify novel biomarkers for early detection of cancer.